Innovation in Software

There’s a lot of interesting things happening out there right now. HTML 5 is about to make a whole suite of new applications possible thanks to:

• Much better rendering of graphics on the fly
• Client-side storage of application data
• Drag-and-drop interfaces that make web apps feel more desktop-like

But there has to be something we are missing out on that is niggling us all at the back of our collective group-think mind. Perhaps watching the Google Wave introduction got you psyched about something that suddenly became possible. Perhaps the very way the web inherently works bothers you, and you envisage a new platform.

I’m interested in hearing about it now the comments are getting a little bit of love across articles. Go crazy. Throw them in there…

Whilst I’ve worked in data centres before – and am all too familiar with how hot, noisy, industrial and dangerous they can be – I sometimes forget how the software industry I now work in has an industrial footprint in those rooms. It’s easy to think of my business as being ‘clean’, because the dirt is so well hidden.

Plans for Google’s new data centre in Dalles, as the blueprints published by Harper’s shows, should remind us just how industrial our business really is.

Combined with the annotation by Ginger Strand, we get a picture of how big this data centre is. Three buildings of over 68,000 square foot each and electricity consumption equivalent to that needed to power 82,000 homes, a third of which will be used just to keep the building temperature at a reasonable level.

Thanks to its location much of the energy used every day will be supplied via hydroelectric power, however its very existence has caused other technology firms to up their data centre spending, and it’s unlikely all of that capacity will be run on renewable power. And besides, every watt of clean energy powering a server is a watt not powering a domestic home.

It’s also worth remembering this isn’t “the” Google data centre. It’s “a” Google data centre.

For years now they have been pushing racks into peering sites and DCs around the globe as well as smaller facilities of their own – an estimated million servers are out there running Google sites, and there are more data centres planned by Google and their competitors over the next four years. Already data centres consume more power in the United States than the army of some 100-million-plus American monster-sized televisions. As the magazine itself says, the Web “is no ethereal store of ideas, shimmering over our heads like the aurora borealis. It is a new heavy industry, an energy glutton that is only growing hungrier.”

Better virtualisation of servers is going to help, but there’s a limit to how much you can virtualise. Is the time now right for us to get smarter again about how we use clock cycles? Is the efficiency-first stance of programming we’ve consigned to the era of the 8-bit machine now going to become fashionable again?

Maybe though, we could do a little to educate the public to make use of this vast industry a little more efficiently. Does the quest for the top 100 current hot trends at Google really suggest that we’re using this power wisely?

Via RoughType

Service Level Agreements are a must-have for Enterprise clients and it has surprised me that so few web companies have used them as a route to making money: if you don’t need an SLA, take the app for free. If you do want an SLA (because say your entire email operation is running on our web service, say), then you need to pony up some cash. It’s worked in open source, so I think it’s a no-brainer for an industry that is service-orientated at its core.

Good news then that Amazon S3 has today announced an SLA which means if they drop below 99.9% uptime per month you can have some cash back. You get even more money back if they drop below 99% uptime. They also agree to give you 60 days notice if they want to get rid of you for any reason – but don’t have to give that reason.

It’s a step in the right direction, but they could make even more money by offering even better SLAs if customers are prepared to spend more money to get them. That money would be capital Amazon would be free to invest in infrastructure which not only enhances S3, but Amazon’s core systems and business.

Do you remember when, as a geek, you thought Eric Schmidt was cool? Maybe it was about the time that he was working at Xerox PARC. Maybe it was when he was working for Bell or Zilog. Maybe it was when he led the development of Java – man, actually, he deserves a slap for that one – or when he moved to Novell. In recent months, his straddling of Google and Apple has been analysed in some depth around the blogosphere.

The problem is, as a businessman, his strategy is all wrong. If you’re a Google shareholder, you should be asking quite serious questions about whether he really understands the economics of software development in a consumer market correctly.

The “road to Damascus moment” for me was when I read that he thinks mobile phones should be paid for by advertisers and that Google will be lining up to deliver the advertising platform needed. This is an idiotic business model. Thinking about it, I now realise that their entire business model is flawed and can be defeated by somebody offering equivalent execution of a product/service but with a low-cost subscription in lieu of advertising. Just because something is free, it doesn’t mean people will want it if it comes with adverts: if you don’t believe me, ask ITV.

The problem with advertising as an economic model behind any business, is that it requires you, the advertiser or publisher, to distract your customer/target/victim. If I build a web-mail system and say “hey, this will let you get your mail done in no time at all!” don’t you think I would be clearly in a position of hypocrisy if my business model required me to distract you constantly with “partner offers” down the side of your inbox?

If I offer you a ‘free’ spreadsheet or word processing tool, don’t you think I might be harming your productivity just a tad by hoping you won’t mind me looking at what you’re working on and making some useful suggestions for websites you should go and visit that want to sell you stuff? You thought clippy the paperclip was annoying? Wait until Google start running ads inside Writely.

The issue for me, is that advertising in the UI is anathema to good software design. We should be aiming to produce high-quality, bug-free software that lets users do the right thing as quickly as possible, and make it difficult for them to do the wrong thing at all. If we accept that the way this development will be paid for is through advertising, we are telling the user “actually, we don’t care about your productivity as much as we should”.

Advertising on mobile phones would be even worse. I look at my mobile phone maybe a dozen times a day. When I use it to browse something on the web, I make sure it’s as quick and painless and ad-free as possible. It is a tool, a device I carry around because I need to. It’s not my best friend that I feel the need to consult whenever I don’t know what to do with my evening.

If people are going to give me a £250 phone for free just so I can watch their ads, they’re going to quickly get burnt and end up having to cut costs and reduce their quality of service to me. They’ll do this as they realise I’m not going to be interrupted by them all day, and their maths of how excited I would be to hear about their special offer collapse under the weight of my considerable apathy.

What I’d prefer is what I have – a contract for 12 months where I give O2 my £25/month, and they give me a free/subsidised phone every year and a few free minutes of calls chucked in as well. I pay this for them to provide me a high quality of service, and to not send me adverts. To be there when I need them to be, and to stay out of my way when I’m, you know, living my life.

I think a lot of customers are going to start buying into this as well – they realise the problems with advertising supported businesses because it hurts their productivity, and when somebody produces a webmail tool as well designed as Google’s, but costing a small subscription year with no ads, Google will see market share slide. People will start to realise that whilst money is expensive and scarce, compared to time it’s cheap and freely available. If you ask somebody “do you want free, but takes longer, or a small amount of money and it’s less time with fewer distractions?” when they want to get something done, they’ll reach for their wallet every time.

The result will be the only people using Google and other ad-supported software being those who can’t pay a subscription, or use the tools so infrequently a subscription is of little value to them: exactly the wrong demographic advertisers want to talk to. As software developers, it is in our economic interest to notice this before Google and others do.

There is another scenario as well. Thinking about this, I came to the conclusion I don’t use Gmail because it is ad supported and I want my mail client to be a productivity tool as much as anything else – I don’t want to be distracted. But I do use Google search. Why don’t I get distracted there? And the answer hit me: my brain filters out the ads. I now just don’t see the links in the blue box at the top, or the boxes down the side of the page: I have adapted to focus on what I need. This is an interesting observation from the perspective of how the human brain adapts, but the scary question for Google is: what does this mean for an advertising-funded business model when everybody adapts?